The present application is directed to piezoelectric material production and more particularly to a process for manufacturing piezoelectric thick film elements and arrays of elements, and structures incorporating such elements.
Piezoelectric ceramic films, e.g., lead zirconate-lead titanate (PZT) and its modified forms are generally defined as being either thin-film elements, up to approximately 10 μm in thickness, or thick-film elements, being approximately greater than 10 μm in thickness. Thin-film piezoelectric elements and thick-film piezoelectric elements greater than approximately 10 μm thick, can be used in a wide variety of applications, including but not limited to microelectromechanical systems (MEMS), microfluid pumps or ejectors, such as jet printheads or acoustic ejectors, and ultrasonic transducers.
Unfortunately, elements in the range of greater than 10 μm to 100 μm are not now able to be produced in high volume with economical yields which permit commercialization. Rather, current methods to make the films in such thickness range are either by polishing the bulk ceramic pieces from more than 100 μm down to the required thickness or using a sol-gel hybrid (or composite) process. The first method is a time-consuming and expensive process which does not lend itself to the making of patterns or arrays. The thick films obtained by the second method have very low quality, are difficult to be patterned, and the required annealing step at 500 to 700° C. limits the substrates which may be used. Thus, there are no cost-effective methods to make high-quality, thick film (greater than 10 to 100 μm) individual elements and arrays, with the elements having arbitrary shapes and on any kind of substrate including silicon, metal and plastics or epoxies.
For many of these applications, the so called thick films, with the thickness range from greater than 10 to 100 μm, are considered beneficial in order to generate a large displacement, apply a large force, to provide a suitable working frequency ranges, and to optimize the performance of actuation or sensing systems. For example, in an existing piezoelectric inkjet printhead, with a stainless steel diaphragm having a thickness of 25 to 40 μm, the thickness of the piezoelectric elements should be about 40 to 70 μm for an optimized design.
Piezoelectric films with the thickness range of greater than 10 to 100 μm are also useful for high frequency (20 to 200 MHz) transducers and catheters used in imaging, such as imaging of arterial walls, structures in the anterior chamber of the eye, and intravascular ultrasound imaging.
These applications may find use for both single element transducers and transducer arrays. For these applications it may be useful to provide the piezoelectric films on polymers, such as some epoxies, which works as backside materials to absorb or diminish backside ultrasonic waves for better image quality, or other advantages.
However, to fabricate piezoelectric films in a greater than 10 to 100 μm thickness range on, suitable substrates for such uses is very difficult for current thin and thick film processes. This is because, the traditional thin film processes, such as sol-gel processing, sputtering and chemical vapor deposition, can only practically generate films with thickness up to 10 μm range. It is also not efficient to use these thin film processes to produce thick films even if they could do so. On the other hand, the traditional thick film processes, such as screen printing, can produce thick films only on the substrates which can withstand higher than 1100° C. temperatures because the screen printed films have to be sintered at about 1100 to 1350° C. for densification and to get good properties.
While a sol-gel hybrid (or composite) method, in which ceramic powders are suspended in a sol-gel solution for spin coating, has been developed at Queen's University of Canada to prepare 0-3 ceramic (powders)/ceramic (sol-gel matrix) composite films with the thickness of 10 to 80 μm on silicon and metal substrates, there are still several drawbacks for this method. First, the film density, and hence the film quality is very low because of low densification process and no grain growth of powders during sintering. Secondly, the film is very difficult to etch or pattern due to its inhomogeneous nature in micrometer scale. Thirdly, as the films have to be sintered at 600 to 700° C., this method can not be used to deposit films on polymers or other substrates which can not withstand 600° C. or higher.
U.S. Pat. No. 6,071,795 to Cheung et al. provides a method of separating a thin film of gallium nitride (GaN) epitaxially grown on a sapphire substrate. The thin film is bonded to an acceptor substrate, and the sapphire substrate is irradiated by a radiation source (such as a laser or other appropriate device) with a beam at a wavelength at which sapphire is transparent but the GaN is strongly absorbing, e.g., 248 nm. After the irradiation, the sample is heated above the melting point of gallium (Ga), i.e., above 30° C., and the acceptor substrate and the attached GaN thin film are removed from the sapphire growth substrate. It was noted that at about 400 mJ/cm2, one pulse of the laser was sufficient to separate the epitaxially grown film of GaN from the sapphire substrate. It is also noted in a specific embodiment, the thin film of the GaN is grown to a thickness of 3 μm.
It is considered that the high energy levels required for the separation process of the thin film GaN, is in part due to the fact that the GaN is epitaxially grown on the substrate, resulting in a degree of lattice matching between the GaN film and the sapphire substrate. This relationship results in a strong adhesive energy between the substrate and GaN.
It is therefore deemed desirable to develop a process which can effectively deposit greater than 10 to 100 μm-thick piezoelectric films on various substrates (silicon, metals, polymers), where the films can be easily patterned during the process, and can produce identical, large-quantity, high-quality thick film elements detachable from the substrate.